Preface for the special issue on Seismic Structural Health Monitoring for the Bulletin of Earthquake Engineering

In the last thirty years, Seismic Structural Health Monitoring (S²HM) has received a growing interest from both researchers and professionals, and an increasingly high number of seismic monitoring systems are currently installed in various countries. One of the main reasons behind this development is connected to the limitations of the traditional methods for condition assessment mainly based on visual inspections. However, an important role is also played by the great potential offered by modern data acquisition, transmission, and processing technologies. They allow a prompt and continuous availability of the information needed to support decisions in emergency situations after an earthquake, and are also useful for maintenance purposes in operational conditions. The aim of this special issue is to report recent advances and successful applications in this field. The volume includes 16 papers from different research groups worldwide and focused on various aspects of S²HM.

The papers by Ramirez et al., Siringoringo & Fujino, Oliveira et al., and Aytulun and Soyoz benefit from data collected on monitored structures to investigate significant aspects of their performance. Environmental sources can strongly affect the structural performance and thereby also the values of the damage indicators. This may lead to false or missing indications of damage if these sources are not duly accounted for. In their paper Ramirez at al. evaluate the daily and seasonal variability of natural frequencies with changes in temperature or humidity on the natural frequencies of three real large-scale structures, a multistorey building, and two sixteenth-century adobe masonry churches. Siringoringo & Fujino discuss the performance of the Tokachi cable-stayed bridge, equipped with moveable spherical frictional plates providing interesting insights into the non-linear damping characteristics of the bridge. The automated processing of seismic monitoring data to extract modal parameters of two large dams and study their evolution in time with varying water levels is the focus of the paper by Oliveira et al. Aytulun and Soyoz report on the investigation of three tall buildings in Istanbul that are monitored continuously, thus capturing their behavior and performance during and after an Mw 5.7 earthquake.

One of the most important and challenging goals of S²HM is the rapid assessment of the structural state and performance in the aftermath of an earthquake. Vibration-based damage identification techniques are thus an important decision support tool that is being investigated by several research groups using both data-driven and model-based approaches. The rapid identification of damage requires data processing methods with a low computational effort. This somehow limits the applicability of model-updating techniques that often require intensive computational effort. However, models can still enable to building functional relationships between the damage indicators and the structural damage level. This is the approach followed in the paper by Sivori et al. in which a seismic damage assessment methodology is proposed and applied to a masonry building. ‘Behavioral charts’ obtained in the pre-event phase through nonlinear analysis correlate natural frequencies to specific levels of structural damage and are used, in the aftermath of an earthquake, to quickly identify the level of damage based on the values of the experimental natural frequencies. A simplified modeling approach is also proposed in the paper by Acunzo et al. to increase the spatial resolution of information through a virtual sensing approach. The response at virtual sensors is built using the response to vibrations measured by the physical sensors and a predictive model identified using the response to vibration in ordinary conditions. Li et al. tackle the subject of optimization of a limited number of sensors using a Deep Neural Network trained using structural displacements. A similar topic is addressed in the paper by Hoult through a method for determining the displacement profile using minimal instrumentation. A different approach to damage identification, which envisages the sole use of data, is proposed and demonstrated using experimental data collected during cyclic and dynamic tests on full-scale RC columns in the paper by Shan et al.

The identification of the modal parameters to characterize the structural behavior is another important goal of S²HM. However, the amplitude-dependent stiffness may limit in some cases the use of operational modal analysis. Martakis et al. tackle this problem by applying an identification Bayesian updating framework using data gathered during demolition activities on 9 masonry buildings. The identification of the non-linear structural performance using seismic monitoring information is the topic of the paper by Yang et al. who propose and apply it to the case study of a reinforced concrete two-story frame, a modified Kalman filter approach.

Technological developments increase the effectiveness of monitoring information as a support tool for rapid decisions. Real-time processing and model updating in monitored systems require sensor networks with low latency and reliable communication. In the paper by Gattulli et al. these aspects are investigated considering the use of “Digital Twins” to improve early warning through the use of 5G transmission protocol, a highly adaptable sensor board, and a 5G Multi-Access Edge. Another important aspect for S²HM, not often investigated, is related to the impact of the quality of data on damage assessment results. Therefore, the paper Liao et al. shows that medium-quality accelerometers (MEMS-based instruments) can provide feasible solutions for cost-effective city-scale deployment and performances that are superior to those of sensor networks with high-quality accelerometers.

The cost-effectiveness of S²HM systems is indeed another important and current research topic that is attracting more and more interest from both researchers and the industry. The investigation of this topic requires a proper quantification of the costs related to different alternatives entailed by the decision problem supported by S²HM information. Besides the direct costs of disruptive earthquakes, related to casualties, fatalities, and damages, the loss of functionality may cause high indirect costs due for example to shutdowns or to non-necessary prolonged downtime of critical facilities. The increased knowledge provided by S²HM can improve decisions thereby reducing these costs. In the paper by Giordano et al. a framework to quantify the value of the S²HM information for the management of seismic emergencies and the selection of optimal actions is proposed. Skolnik et al. discuss examples of commercially available solutions to reduce indirect costs through monitoring information.

This collection of 15 papers is the result of the efforts of several research groups worldwide. We wholeheartedly thank all authors who contributed and also many colleagues involved in the review process. We hope the readers will find this selection of papers informative and interesting at least as much as we did.